DETAILED ACTION
This Office Action is sent in response to Applicant’s Communication received 9/20/2023 for application number 18/471,196.
Claims 1-20 are pending.
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Claim Rejections - 35 USC § 102
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
Claim(s) 1- is/are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Caccia et al., SPeCiaL: Self-supervised Pretraining for Continual Learning (NPL [U], see Notice of References Cited).
In reference to claim 1, Caccia discloses a computer-implemented method for performing meta-learning (Abstract, page 91), the method comprising: performing a first set of training iterations to convert a prediction learning network into a first trained prediction learning network (model “head” is the prediction learning network, see page 95, bottom to page 96, top) based on a first support set of training data (during pretraining, the model head is trained I the inner loop using a support set, see page 95 and fig. 1 on page 93); executing a representation learning network (model “base” / “trunk” is the representation learning network, see page 95, bottom to page 96, top) and the first trained prediction learning network to generate a first set of supervised training output and a first set of self-supervised training output based on a first query set of training data corresponding to the first support set of training data (in outer loop, using a query set with samples from support set is sent through base and head, page 95; the sets are both supervised and self-supervised because the data includes both labeled data and augmented data, 4.1 Generating Pseudo Labels, page 96, fig. 1 on page 93); and performing a first training iteration to convert the representation learning network into a first trained representation learning network based on a first loss associated with the first set of supervised training output and a second loss associated with the first set of self-supervised training output (model base learns representations based on loss from both the labeled and augmented data, page 95, Pretraining, pages 97-98), wherein, in operation, the first trained representation learning network generates a latent representation of a data sample that is not associated with a set of labels for the first query set of training data (in deployment, the model base, or representation network, extracts representation, Deployment: Few Shot Continual Learning, page 98; 1. Introduction, pages 91-92).
In reference to claim 2, Caccia discloses the computer-implemented method of claim 1, further comprising performing a plurality of training operations on the first trained prediction learning network using the first loss to generate a second trained prediction learning network (loss from labeled examples is backpropagated through inner learning procedure to further train “head” or prediction learning network, page 95).
In reference to claim 3, Caccia discloses the computer-implemented method of claim 2, further comprising: performing a second set of training iterations to convert the second trained prediction learning network into a third trained prediction learning network; and performing a second training iteration to convert the first trained representation learning network into a second trained representation learning network based on additional training output generated by the first trained representation learning network and the third trained prediction learning network (above process is iterated for further sets of data until convergence, page 95).
In reference to claim 4, Caccia discloses the computer-implemented method of claim 2, further comprising: performing a second set of training iterations to convert the second trained prediction learning network into a third trained prediction learning network based on a second support set of training data; and executing the first trained representation learning network and the third trained prediction learning network to generate one or more predictions associated with a second query set of training data corresponding to the second support set of training data (above process is iterated for further sets of data until convergence, page 95).
In reference to claim 5, Caccia discloses the computer-implemented method of claim 1, wherein performing the first set of training iterations comprises: applying, during a first iteration included in the first set of training iterations, a first training update to the prediction learning network based on a first training sample included in the first support set of training data; and applying, during a second iteration included in the first set of training iterations, a second training update to the prediction learning network based on a second training sample included in the first support set of training data (during inner loop training of “head,” or prediction network, the prediction network sequentially takes samples from set and is updated at each step, page 95).
In reference to claim 6, Caccia discloses the computer-implemented method of claim 1, wherein performing the first set of training iterations comprises: executing the representation learning network and the prediction learning network based on the first support set of training data to generate a second set of supervised training output and a second set of self-supervised training output; performing a plurality of training operations on the prediction learning network using a third loss associated with the second set of supervised training output; and performing a plurality of training operations on the representation learning network using a fourth loss associated with the second set of self-supervised training output (during inner loop training of “head,” or prediction network, both labeled and transformed data are sent through base and head, i.e. representation and prediction networks, to get supervised loss from labeled data and self-supervised loss from transformed data, see e.g. fig. 1 on page 93, and both losses are backpropagated to update head, page 95).
In reference to claim 7, Caccia discloses the computer-implemented method of claim 1, wherein executing the representation learning network and the first trained prediction learning network comprises: executing the representation learning network to convert the first support set of training data into a first set of latent representations; and executing the first trained prediction learning network to convert the first set of latent representations into the first set of supervised training output (outer loop training executes base to get representations, then sends representations trhough head to get output, fig. 1 on page 93, page 95).
In reference to claim 10, Caccia discloses the computer-implemented method of claim 1, wherein the first set of supervised training output comprises a prediction of a label for an image included in the first query set of training data (image labels, fig. 1 on page 93).
In reference to claim 11, this claim is directed to a non-transitory computer-readable medium associated with the method claimed in claim 1 and is therefore rejected under a similar rationale.
In reference to claim 12, Caccia discloses the one or more non-transitory computer-readable media of claim 11, wherein the instructions further cause the one or more processors to perform the step of performing a second set of training iterations to pre-train the representation learning network based on the first support set of training data and a second set of self-supervised training output (training iterations can be pre-training, page 95).
In reference to claim 13, Caccia discloses the one or more non-transitory computer-readable media of claim 12, wherein the second set of training iterations is performed before the first set of training iterations (pretraining has a plurality of iterations, so some iterations are before others, page 95).
In reference to claim 14, Caccia discloses the one or more non-transitory computer-readable media of claim 11, wherein the instructions further cause the one or more processors to perform the steps of: performing one or more training operations on the first trained prediction learning network using the first loss to generate a second trained prediction learning network (during inner loop training of “head,” or prediction network, the prediction network sequentially takes samples from set and is updated at each step, page 95); and executing the first trained representation learning network and the second trained prediction learning network to generate one or more predictions associated with a second query set of training data (outer loops further iterates over additional query sets, fig. 1 on page 93, page 95).
In reference to claim 15, Caccia discloses the one or more non-transitory computer-readable media of claim 14, wherein the instructions further cause the one or more processors to perform the steps of: performing a second set of training iterations to convert the second trained prediction learning network into a third trained prediction learning network based on a second support set of training data; and executing the first trained representation learning network and the third trained prediction learning network to generate one or more predictions associated with a second query set of training data corresponding to the second support set of training data (above process is iterated for further sets of data until convergence, page 95).
In reference to claim 16, this claim is directed to a non-transitory computer-readable medium associated with the method claimed in claim 5 and is therefore rejected under a similar rationale.
In reference to claim 17, Caccia discloses the one or more non-transitory computer-readable media of claim 11, wherein performing the first training iteration comprises: applying a first training update to the representation learning network based on the second loss associated with the first set of self-supervised training output; and applying a second training update to the representation learning network based on the first loss associated with the first set of supervised training output (representation learning network is trained over a plurality of iterations, page 95).
In reference to claim 18, Caccia discloses the one or more non-transitory computer-readable media of claim 11, wherein the first loss comprises a classification loss between the first set of supervised training output and the set of labels included in the first query set of training data (supervised loss based on labels, page 95, fig. 1 on page 93).
In reference to claim 20, this claim is directed to a system associated with the method claimed in claim 1 and is therefore rejected under a similar rationale.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claim(s) 8-9 is/are rejected under 35 U.S.C. 103 as being unpatentable over Caccia et al., SPeCiaL: Self-supervised Pretraining for Continual Learning (NPL [U], see Notice of References Cited) in view of Pathak et al., Context Encoders: Feature Learning by Inpainting (NPL [V], see Notice of References Cited).
In reference to claim 8, Caccia does not explicitly teach the computer-implemented method of claim 1, wherein executing the representation learning network comprises: executing a first portion of the representation learning network to convert the first support set of training data into a first set of latent representations; and executing a second portion of the representation learning network to convert the first set of latent representations into the first set of self-supervised training output.
Pathak teaches the computer-implemented method of claim 1, wherein executing the representation learning network comprises: executing a first portion of the representation learning network to convert the first support set of training data into a first set of latent representations; and executing a second portion of the representation learning network to convert the first set of latent representations into the first set of self-supervised training output (feature representations are learned by an encoder taking an image with missing region and a decoder filing in the missing region, see 3. Context encoders for image generation on pages 2538-41).
It would have been obvious to one of ordinary skill in art, having the teachings of Caccia and Pathak before the earliest effective filing date, to modify the representation learning network of Caccia to include the self-supervised training of Pathak.
One of ordinary skill in the art would have been motivated to modify the representation learning network of Caccia to include the self-supervised training of Pathak because it helps learn more meaningful image features (Pathak, 1. Introduction, page 2537).
In reference to claim 9, Caccia does not explicitly teach the computer-implemented method of claim 1, wherein the first set of self-supervised training output comprises an infilling result associated with an image included in the first query set of training data.
Pathak teaches the computer-implemented method of claim 1, wherein the first set of self-supervised training output comprises an infilling result associated with an image included in the first query set of training data (feature representations are learned by an encoder taking an image with missing region and a decoder filing in the missing region, see 3. Context encoders for image generation on pages 2538-41).
It would have been obvious to one of ordinary skill in art, having the teachings of Caccia and Pathak before the earliest effective filing date, to modify the representation learning network of Caccia to include the self-supervised training of Pathak.
One of ordinary skill in the art would have been motivated to modify the representation learning network of Caccia to include the self-supervised training of Pathak because it helps learn more meaningful image features (Pathak, 1. Introduction, page 2537).
Claim(s) 19 is/are rejected under 35 U.S.C. 103 as being unpatentable over Caccia et al., SPeCiaL: Self-supervised Pretraining for Continual Learning (NPL [U], see Notice of References Cited) in view of Sun et al. (US 2021/0158512 A1).
In reference to claim 19, Caccia does not explicitly teach the one or more non-transitory computer-readable media of claim 11, wherein the second loss comprises a mean squared error between the first set of self-supervised training output and a set of data samples included in the first query set of training data
Sun teaches the one or more non-transitory computer-readable media of claim 11, wherein the second loss comprises a mean squared error between the first set of self-supervised training output and a set of data samples included in the first query set of training data (loss can be based on mean squared error, para. 0025, 0039).
It would have been obvious to one of ordinary skill in art, having the teachings of Caccia and Sun before the earliest effective filing date, to modify the loss of Caccia to include the MSE of Sun.
One of ordinary skill in the art would have been motivated to modify the loss of Caccia to include the MSE of Sun because MSE is a common loss function.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Andrew T. Chiusano whose telephone number is (571)272-5231. The examiner can normally be reached M-F, 10am-6pm.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Tamara Kyle can be reached at 571-272-4241. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/ANDREW T CHIUSANO/ Primary Examiner, Art Unit 2144
Prosecution Insights